After learning the secret of “tonsiliths”, my significant other is beginning to appreciate TSD more. (“Well why don’t you go ask CECILL…”)
Very well, I shall. Last night, we were playing with keychain lights at an aquaintance’s house and one posed the question: “Why are LED’s different colors?”
I heard them say some things about frequency and vibration, but that sort of technology is far beyond me.
These seemingly identical clear bulbs range in color from red to blue to green etc. What gives?
Inside an LED is a block of material made from two slightly different semiconductors. When in use electricity flow from one wire onto one side of the block, across the gap onto the other, and out onto the other wire.
Now, in elements, the outer valence electrons (which are what’s involved in carrying urrent flows) exist at different distances from the nucleus, depending on the material.
Because the two different sides of the semiconductor block are made from different materials, there is an energy difference in the electron orbital levels from one side to the other. Electrons flowing from one side to the other act like a marble roling across the floor and meeting a step - they “fall” from the higher energy level material to the lower energy level material, and give off a “thump” in the form of a photon of light as they do.
The energy level of the photon of light emitted is equal to the diference in orbital energy levels between the two materials. Materials with very similar electron orbitals will give off low-energy photos, of red or infrared light. Greater differences in electron orbitals between the two materials gives higher-energy photons, of green or blue light.
So, it’s simply a matter of the materials which the diode junction is made from. Low energy junctions were easiest to make, which is why red and infrared lights came first. Finding the right materials to make a blue LED took a lot longer, and they’re still a lot more expensive than the common red types. There are about a dozen different semiconductor juction types in production now, each with its own emission wavelength.
Great answer, Andrew! I have also wondered about this and now I know.
Since I know very little about the purpose of electronic components I’m going to probably ask some “dumb” questions. But, from what you are saying, the materials have to be of different composition at the junction to get this “thump”, correct? The junction can’t be of the same material or nothing will happen? Is that what diode is supposed to do, boost/release energy before it reaches a particular circuit? Do LED’s ever wear out?
The junction has to be of different materials or it has no functions. Semiconductor materials are doped with specific impurities named N and P types. It’s been 20 years since I had basic semiconductors but IIRC (and please correct me someone if I goof up) electrons flow readily from N to P across a junction but not the other way under normal circumstances. A single junction makes a diode which is nothing more than a one way valve which has a lot of uses in electronics. Put too much voltage the “wrong” way and the junction breaks down. Zener diodes are made to do this at a specific voltage and are used for regulators. Put three semiconductors together with two junctions and a wire lead to each and you have a transistor which can work as a variable valve.
By definition, a diode (whether vacuum tube or semiconductor) is an electronic component with two electrodes. I diode’s simplest function is to allow current to flow in only one direction. This is what allows it to change alternating current into pulsating direct current. The diode is therefore the basic (but not the only) component of DC power supplies.
Light emitting diodes behave in a similar manner, but their unidirectional properties are not what LEDs are prized for. In order for current to begin flowing through a diode junction, a certain voltage threshold must be exceeded. This leads to the thump that I suspect Andrew was talking about (although I’ve never heard that word used to describe the operation of electrons moving across a P-N junction). For regular (non-LED) silicon diodes, the threshold is 6-tenths of a volt.
LEDs eventually wear out, but in all probability the device the LED is part of will wear our long before the LED does, so we never really hear about LEDs wearing out. I’ve worked on some old consumer electronics & seen some LEDs with weak output, but it was never so low that I felt replacing the LED was called for. Of course some LEDs will go bad prematurely due to faulty manufacturing, as will anything.
Here is a nifty chart that details the types of materials needed in order to get certain colors. Also listed is the voltage threshold that I mentioned earlier. And here is a see-through picture of a diode allowing you to see the naughty bits inside.
Keep hitting the “next” links for more info on LEDs, including the new fangles bright white light LEDs that may soon replace incandescent bulbs.
I hope that’s not too technical for a Sunday morning. 
:Checks clock:
Or a Sunday afternoon.
Finally being able to make green LED’s was a big deal. It required a lot of fiddling with band gap energies and so on. Since they already had blue and red, this means you can set them up in the standard red-green-blue configuration found in TV’s, etc. Future plans include making flat screens with LED’s rather than LCD’s and making green laser pointers. That’s because the human eye is much more sensitive to green light than red light. Consequently, you don’t need to have as powerful of a laser, which minimizes the chance that some punk shining one in your eye will cause permanent damage.
Thank you, tour of 3M. I hope I didn’t give away any trade secrets.